Descriptive Criticism

Nearly every teacher of any kind of art has, at one time or another, engaged her or his students in peer criticism.  Even if that hasn't happened, the teacher has certainly put forth "constructive criticism" him or herself.  But I'm not sure we always know quite what we mean by this, and I'm even less sure that we always go about the process of criticizing students' work in the best way.  I'm going to outline here a particular approach to both peer and teacher criticism--one that I try to confine myself to as a teacher, and one that I require my students to confine themselves to when critiquing peers' work.  I first encountered this approach as a graduate student in playwriting for youth, under the mentorship of Dr. Suzan Zeder at the University of Texas.  It is a philosophy that can be put to use in any medium.  Without presuming to tell anyone that they should adopt this approach to the exclusion of all others, let me tell you why I mostly have done so.


The idea starts with two basic principles:


1.)  The purpose of criticism is to help the artist to improve his or her work.


2.)  No one but the artist can every truly know the artist's intent.


Let me elaborate a little on each point.


1.)  The purpose of criticism is to help the artist to improve his or her work.  I don't mean this only in the general sense of "helping him or her to become a better artist--a better painter, a better playwright, a better actor, a better novelist.  I mean helping the artist to improve the particular work being criticized. That's a hugely important distinction.  When I make the choice to expose a play to the criticism of my peers or my mentor, it isn't simply in the hope that I'll learn something that will help me write a better play next time.  I am hoping to learn things that will help me make this play better.  Inherent in this idea, of course, is the idea of revision as an essential part of the creative process, and for this reason the approach works a little better in some media than in others--it's a little difficult to revise a marble statue, for example.  But it still works as a general principle.  Therefore, I want the criticism I receive to be structured in the best way to achieve that end.


2.)  No one but the artist can ever truly know the artist's intent.  When I receive (view, hear, see) a work of art, I can know what the art says to me--what I have received--but I cannot know whether what I receive is what the artist intends me to receive.  That means that any criticism I offer that is predicated, even unconsciously, on my idea of what the artist intends is bound to be flawed and may be actually harmful--especially if, as is usually true, my assumptions about what the artist intends are unconscious.  (Because when I give the criticism, I am unlikely to tell the artist what those assumptions are if I don't consciously know I'm making them.)  In other words, in trying to help the artist to bring the work more in line with what I think it ought to say or what I think it's trying to say, I may be giving advice which, if followed, would actually bring the work further from the artist's intent.  Therefore, I must be very careful to structure my criticism in such a way as to avoid making assumptions about the artist's intent.  (Even if the artist tries to tell me what is intended, this is still a danger, language being often inadequate in this area.)



There are basically two kinds of criticism.  The first, and most familiar, can be called "prescriptive criticism."  This is criticism that, either explicitly or implicitly, tells the artist what to do or what to change.  Very simple examples of this kind of criticism abound in theatre education.  Explicit examples include things like, "speak louder," "you need to make the character's anger more clear," or "I want you to expand Act I."  Remarks like, "the second scene is too short," or "you're going too fast" also fall into this category, because you're implicitly saying "make the second scene longer" or "slow down."  Now, obviously, such comments are sometimes necessary in a fast-paced project like, say, directing a play, but there's an inherent problem in them if you want to really respect that artist's integrity, and that is that every one of them, even the most simple, is ultimately predicated on an assumption about the artist's intent.  When you say, "speak louder," it's because you can't hear the actor clearly and you assume that this is not an intentional choice on the actor's part.  Now, in this example, that's almost always a fairly safe assumption, but that's actually why I chose to begin with such a simple example--because it is at least theoretically possible that the actor intends to be difficult to hear.  Perhaps he wants the audience to lean forward, to listen harder.  Perhaps he wants to make them uncomfortable.  (There have been experimental theatre groups that put deliberately uncomfortable seats in their performance spaces for this reason.)  In the next example, "you need to make the character's anger more clear," the situation is much murkier.  Something in either the rest of the actor's (or writer's) performance or in the text itself has led you to assume either that the artist intends to convey anger (even though it's not coming across) or that the artist should intend to convey anger.  But the artist may not intend that at all--and this is a much more subjective area, where, even if you are an experienced artist and teacher and the artist is as green as grass, you may not know better than the artist where the character ought to go.


The other kind of criticism can be called "descriptive criticism."  At its most pure, this is criticism that seeks to tell the artist not what the critic thinks he or she ought to have received from the work, not what the critic thinks the artist intended to convey, but only what the critic has actually received.  In the first example above, the descriptive comment would be "I couldn't hear you."  In the second, it might be "I wondered why Bob wasn't more upset by Tom's actions."  Simply put, prescriptive criticism prescribes what you ought to do, where descriptive criticism describes what you actually did and leaves it up to you to decide whether you want to make changes.  Why is this better?  Well, first, it should be clear that it can't be worse.  In the first example, if (as is, of course, overwhelmingly likely) the artist does not intend to be inaudible, he will take exactly the same corrective action as he would take if told baldly to increase his volume.  In the second, if the artist does intend to convey Bob's anger, the information that he has failed to do so will, again, produce precisely the same action--some attempt to convey the anger more clearly--as would be produced if he were told specifically to address the issue.  It is the possibility that anger is not the artist's intention that makes the descriptive approach better.  On learning that the audience (or reader, or whatever, depending on the medium) feels some disconnect between Tom's actions and Bob's reaction to them, he will be left to address this disconnect in any way that seems best to him--clarifying the reasons for Tom's actions so that they no longer seem to require an angry response from Bob, clarifying Bob's character so that we can see that he's not the sort to get angry--whichever approach serves his artistic intent.  Moreover, in a well-constructed peer criticism session, if the remark is somehow confusing to the artist--it never having occurred to him that Bob has anything to be angry about, for example--he can ask follow up questions of his own design to try and figure out which part of what he is trying to convey has failed to come across.


It may be worth briefly addressing the idea of communication here.  In all of the above, I'm making two assumptions, both of which are nearly universally true in the theatre, but may be somewhat less true in other media.  First, I'm assuming that there is an intent on the part of the artist to communicate something.  Second, I'm assuming that the artist cares whether his audience (readership, viewers, etc.) understands that communication.  I have met visual artists and composers (and one stage director--the experimental director Robert Wilson) who have baldly said either that their work isn't meant to say anything in particular or that it doesn't matter at all to them what people receive from it--or, indeed, whether they receive anything at all.  I respect this idea, though I don't understand it.  However, if an artist takes that approach, it's not so much an argument against limiting criticism to the descriptive kind as it is an argument against bothering with any kind of criticism.  The artist isn't going to change anything either way.  So, if one takes the view that the purpose of criticism is to help the artist to refine the work, it makes sense to begin with the assumption that it matters what the audience receives from it.  And here's an important point: The artist is responsible for what the audience receives, whether or not it is intended.  If you are misunderstood, it's no good saying it's because the audience is stupid and they just don't get it--if your intention is to communicate something to your audience, it's your job to make it intelligible for them.  A humorous real-life example:  at the university where my uncle used to teach animal husbandry, a sculptor was commissioned to make a large sculpture to sit outside the new genetics building.  The resulting work was intended by the artist as a wholly abstract, decorative piece.  Unfortunately, taken in conjunction with its location outside the genetics building, the sculpture's shape conveyed, to a lot of people, a rather unfortunate part of the anatomy of a bull.  People in this culturally conservative community were offended, and demanded that the piece be removed.  The artist's initial angry reaction could be summed up as, "No, you idiots, that's not what it means at all!  It's not my fault you all have dirty minds!"  It's an understandable reaction, but I believe she was wrong.  (Had the sculpture been made for her own pleasure, or offered up for sale in a gallery, she might have had a point.  Folks who didn't like it wouldn't buy it, but presumably not everyone would be offended by it.  But a commissioned work, like any work of theatre, commissioned or otherwise, must take its expected "audience" into account, and she had created a work that was not acceptable to the "audience" for which it was intended--those in the community who would pass by the genetics building.)  It's never any good blaming the audience if a work fails to communicate its intent.  (Even if the audience does consist of idiots with dirty minds, if that's the audience you want to reach, it's your job to make work that idiots with dirty minds will understand--which is not to say that there's necessarily anything wrong with making a work that's not intended for everyone.  The director of a Merchant/Ivory-type period drama doesn't worry about whether 20-something "fanboy" types will understand the subtle meanings of his compositions, because they're not the ones who will be coming to see the movie.  But the director of the next Marvell Comics adaptation had better care about them.)  I like to put this idea to my students like this:  "If nobody understands what you said, you didn't say it."


So, if one accepts the general premise that descriptive criticism is the most effective (and least harmful) kind, how does one put that idea into practice?  Am I really saying you should never say "speak up," or "slow down?"  Well, yes and no.  When I direct a play, I do sometimes make prescriptive comments.  It's undeniably quicker, and time is nearly always short in any rehearsal process, but there's another distinction.  A director is not simply trying to help the actor improve her performance in a vacuum--she's trying to improve the production, which is, in a sense, the director's own work of art.  When teaching artists, as opposed to directing them, yes I really am saying you should never say "speak up" or "slow down"--not because those particular prescriptive comments are likely to cause any problems, but because it's a healthy habit to get into--otherwise you're almost bound to make other prescriptive comments that will.  In particular, in peer criticism sessions I insist that my students avoid prescriptive comments, and when one makes one--even if it's as innocent as "you need to talk louder" I stop them and make them rephrase the comment in a descriptive way.  (By the way, if the students are mature enough to grasp the concept, I also forbid general comments like "It was really good."  Strictly speaking, if one accepts the first premise above, that the purpose of criticism is to help the artist improve, such remarks are not criticism at all.  Clearly they're completely useless.  Even adults in peer criticism situations tend to begin their remarks, "Well, first of all, I though it was really good overall."  As an artist, I just don't need to hear that.  I feel like saying, "Well, of course it's good--I'm a good writer!  But I wouldn't be here if I did not think it could be even better, so stop wasting my time and give me something I can use."  Anyway, as an artist, I'm much less interested in whether you like my work than in whether you get it.  Obviously, negative remarks in the same vein are equally pointless.)  However, while I do think it's good to get into the descriptive habit in general, where the concept is most useful is in designing criticism sessions.


So what does it mean to "design" a criticism session?  What's to design?  You show the work and invite comment, right?  Well, no--not if you seriously want to improve the work.  The best peer criticism sessions are highly structured and carefully moderated.  I'll give you an example of how this might work.  Since my main work as a creative artist is in playwriting, I'll describe a peer criticism session for a new play, but the general concept works with any medium.  it starts with careful preparation.  When I know a work is shortly going to be read (for example, to a group of fellow playwrights in the collective to which I belong), I start by asking myself what I'm hoping to say with the work.  I come up with a list of specific questions about elements of the plot, characters, themes, etc. that I want answered.  I am very careful to frame these questions in non-leading ways.  For example, in my play, Ernie's Place, there's a bit of "visual text" (my term for nonspeaking business that is meant to convey meaning) in which 12-year old Stella pauses on her way out of the "super secret clubhouse" to adjust a sign tacked to it, "you must be this short to enter."  My intent with this moment was to convey Stella's growing uneasiness with the fact that she is growing out of her childhood, and, more generally, with the general impermanence of structures in her life.  But I would most certainly not ask, "did you understand that moving the sign is Stella's way of staving off her impending adulthood?"  Instead, I would ask, "why did Stella adjust the sign?"  That way I get answers that are unbiased by information that's not actually in the play.  (If I asked the question the other way, a viewer might well think "Oh--yeah, that makes sense," and answer in the affirmative even though he hadn't actually thought any such thing until I asked the question--and he would likely be unaware himself that he hadn't seen it all along--memory works that way.  Once a thing is obvious, we tend to remember that it was always obvious.)  Once I have compiled a list of questions, I will often print them and hand them out after the reading, before any discussion has taken place.  (This works better than simply posing the questions out loud to the group, because in the second scenario, once one audience member has articulated an idea, the others are quite likely to absorb it unconsciously.  Thus, if no one understood my intent, I'll learn that fact, but if even one person did, I may never learn that only one in ten did.)  If the structure of the reading event will allow it--say if it's not in any way a public performance--I might even stop the reading at one or more points in the middle to hand out questions.  It can be very useful to learn what an audience is thinking at various points prior to the conclusion.  (This is why I always spend several performances of a new children's play sitting down front, script in hand, watching the audience and taking notes--but that's really only effective with audiences of children, who don't censor their reactions the way adults do.)  Only after folks have filled in and handed back the written questionnaires does any discussion begin.


Even then, the discussion is carefully moderated--and not just the "audience's" part of it.  Just as the peer critics must be careful to avoid telling the author what they think she should have done, or what they themselves would have done in her place (which amounts to the same thing), the author must be careful not to invite prescriptive comments by saying things like, "how would you resolve the situation?"   (I'm not suggesting that an artist should never solicit concrete suggestions from a fellow artist.  Particularly in the theatre, which is an inherently collaborative art form, that would be absurd.  But I tend not to allow it to happen during formal structured critiques because once it gets its toe in the door, prescriptive criticism is very difficult to evict.)  One way to ensure that you get descriptive criticism in such a session (though it's counterintuitive at first) is to allow only questions from the audience.  It's impossible to phrase an explicitly prescriptive comment as a question.  (It is not impossible to ask an implicitly prescriptive question, though, such as "Why didn't you show us the argument between them?"  For this reason, I sometimes further restrict the audience to only questions about the actual characters or events in the play, and not about the author.  There's a subtle but very important difference between "Why doesn't she fight back?" and "Why didn't you have her fight back?")  One very important point about taking the "questions only" approach is that the author is not expected to answer the questions.  (This brings up an important point about the purpose of criticism, which I'll take up below.)  There should also be an opportunity for the author to ask questions herself.  Hopefully she's already done so in writing, but she should be ready to ask for clarification if she doesn't understand someone's question or comment.  The moderator (who might be the author herself, though, as I'll say below, it's usually better if it's not) can decide how much to relax the "questions only" rule if it's in place, in order to let the person answer.


All of which brings up an irony.  In seeking to create a setting for descriptive criticism, I'm being awfully darned prescriptive myself!  Dictating what kinds of comments the audience can make, asking the author not to answer the audience's questions, making folks fill out questionnaires--good grief!  This is why it is very important that everyone understands what the purpose of the session is--and it might not be exactly what you want it to be.  For example, when a new work is performed publicly and the audience is invited to a "Q & A" session afterwards with the playwright, they're apt to get pretty testy if he refuses to answer their questions.  And that's perfectly reasonable, because they didn't come to the session primarily to help the playwright hone the script--they came to learn something about the playwriting process, or to express their opinions.  The structure I've outlined above won't work in a situation like this.  Even in a "private" setting, such as a reading where the "audience" consists of fellow playwrights who have been invited by the author, resentments can boil up if the structure is unfamiliar.  No one likes being told what they can and cannot say, and if they view themselves as doing the author a favor by attending, they may like it even less.  But that's not a reason not to do it--it's only a reason to communicate very carefully in advance.  You're basically saying, "Please understand--the whole and only purpose of this reading and discussion is to help me refine my play.  It's not for you to learn about my process, or for your entertainment.  You're doing me a favor--a big one--and the only compensation I can offer you is to promise I'll do the same for you another time."  Which is not to say that you can't, or shouldn't, also participate in events that have a different purpose.  I frequently do Q & A sessions with casts of children who have performed my work, during which I happily answer questions about my process, about what I was trying to say, or about any choices I made.  But I don't view these sessions as for me.  They're for the kids, who want to learn about the theatre and about playwriting.  (Certainly in most cases they're not about refining the script, since it's long since finished and published--which doesn't mean I don't learn anything in the session.)  Similarly, I'll happily talk about my process and my choices with fellow playwrights, formally or informally--but not during a structured criticism session.


The above is the ideal, and very effective when it can be done, but it's often difficult to arrange, and it's entirely impractical for day-to-day use.  (In other words, it's great for a playwright with a whole play, or a whole act ready to be critiqued, or an actor with a polished performance to show, but it doesn't work for the constant critical help an artist needs all through the process--such as when a teacher is coaching a young actor through the process of learning and rehearsing a monologue, and it's not much use in helping with the general development of an artist, as opposed to the development of a particular work.)  Still, if one keeps the general goal of descriptive criticism in mind, it can lead to the development of improved teaching (and assessment) tools in all sorts of ways.  For example, I have discovered that one very powerful form of inherently descriptive criticism is imitation.  Many of my lesson activities involve students "reflecting" their peers' work back to them by imitating it, or by restating it.  (See Three Words in the Lesson Plans section.)  I've also developed assessment tools that strive to assess students with respect to their own artistic intent rather than mine.  (See Artist-Centered Evaluation in the Large Projects section.)  More generally, I've found that by keeping the concept in mind, and by allowing my language to reflect it as much as possible, I am able to make my students much more comfortable with being criticized, and much more open to revision.  A teacher--who clearly did not understand what I was doing--once accused me of being self-centered.  She was picking up on the fact that, in my interactions with the young actors I was coaching, the word "I" predominated, as opposed to the word "you."  It was, or course, intentional.  I was saying things like "I don't understand your motivation here," instead of saying "you're not making your motivation clear."  I was "being the audience," reflecting the actors' performances back to them so that they could make their own choices about how to modify them, rather than telling them what to change.  It's probably exactly the opposite of how, say, a tennis coach might work, carefully prescribing exactly how a young athlete needs to modify his stance, or his grip or whatever--but it's the right way to work with artists.